# batch_compose

Python API: `luxonis_ml.data.augmentations.batch_compose`

## Classes

### BatchCompose

Compose batch-aware Albumentations transforms.

#### Methods

##### init

```python
def __init__(transforms: TransformsSeqType, *, bbox_associations: dict[str, dict[str, str]] | None = None, **kwargs):
```

Compose batch transforms.

Parameters

 * `transforms` (`TransformsSeqType`): Transformations to compose.
 * `bbox_associations` (`dict[str, dict[str, str]] | None`): Bbox fields mapped to their associated target fields and target
   types. Fields listed here must carry the index column appended by
   [luxonis_ml.data.augmentations.utils.preprocess_bboxes](https://docs.luxonis.com/software-v3/ai-inference/model-source/training/luxonis-ml/luxonis-ml-api-reference/data/augmentations/utils.md);
   any other bbox field is left untouched.
 * `**kwargs`: Additional arguments passed to
   [A.Compose](https://github.com/albumentations-team/albumentations/blob/66212d77a44927a29d6a0e81621d3c27afbd929c/albumentations/core/composition.py#L609).

#### Attributes

##### batch_augmentation_indices

##### batch_size

Product of nested transform batch sizes, ∏ibi.

##### transforms

Batch transformations in composition order.